Information Value

Preload

Preload things.

Information Value

Prepare frame

Dictionary:

• numeric: 927 features (Numeric)
• categorical: 227 features (Categorical)
• date: 161 features (Date)
• NA_features: 128 features (d_NA)
• Mean_features: 128 features (d_Mean)
• Min_features: 128 features (d_Min)
• MinMin_features: 1 features (d_MinMin)
• Max_features: 128 features (d_Max)
• MaxMax_features: 1 features (d_MaxMax)
• Time0_features: 128 features (d_Time0)
• FirstLast_features: 2 features (d_First, d_Last)
• Timing_features: 128 features (d_Timing)
• TimeToExitStation_features: 128 features (d_TtoE_S)
• TimeToExitProduction_features: 1 features (d_TtoE_P)

# Frame of scores
ig_frame <- fread("E:/Laurae/NumericCMI_exact_best_grid/IG_2216feat.csv", data.table = FALSE)
ig_frame <- data.frame(Feature = ig_frame$Feature,
                       Type = ig_frame$Type,
                       Levels = ig_frame$Levels,
                       Splits = rep(0, 2216),
                       NotMissing = ig_frame$NotMissing,
                       IG = ig_frame$IG,
                       IV = rep(0, 2216),
                       RankIG = ig_frame$RankIG,
                       RankIV = rep(0, 2216), stringsAsFactors = FALSE,
                       Number = 1:2216)
# Folds for CV
folds <- numeric(1183747)
for (i in 1:5) {
  folds[fread(paste("E:/Laurae/folds/folds0", i, ".csv", sep = ""))$x] <- i
}

Precompute all Information Values

Formulas:

• Mutual Information = I(X, Y) = H(X, Y) - H(X | Y) - H(Y | X), understand un-sided
• Conditional Mutual Information = I(X, Y | Z) = H(Z | X) + H(X) + H(Z | Y) + H(Y) - H(Z | X, Y) - H(X, Y) - H(Z) = I(X, Y) + H(Z | X) - H(Z | Y) - H(Z | X, Y) - H(Z), with guarantee of I(X, Y | Z) > 0 and being one-sided

gc(verbose = TRUE)

Garbage collection 90 = 37+8+45 (level 2) ... 
89.3 Mbytes of cons cells used (63%)
18438.7 Mbytes of vectors used (62%)

             used    (Mb) gc trigger    (Mb)   max used    (Mb)
Ncells    1671949    89.3    2637877   140.9    2637877   140.9
Vcells 2416793867 18438.7 3910687954 29836.2 2420182705 18464.6

large_data <- data.frame(variable = character(),
                         class = character(),
                         pct_bin = numeric(),
                         good = numeric(),
                         bad = numeric(),
                         bad_rate = numeric(),
                         inflation = numeric(),
                         pct_good = numeric(),
                         pct_bad = numeric(),
                         odds = numeric(),
                         woe = numeric(),
                         miv = numeric(),
                         stringsAsFactors = FALSE)
default <- data.frame(variable = character(),
                      class = character(),
                      pct_bin = numeric(),
                      good = numeric(),
                      bad = numeric(),
                      bad_rate = numeric(),
                      inflation = numeric(),
                      pct_good = numeric(),
                      pct_bad = numeric(),
                      odds = numeric(),
                      woe = numeric(),
                      miv = numeric(),
                      stringsAsFactors = FALSE)
# Magic function to coerce super quickly
setDF <- function(x) {
    if (!is.data.table(x))
        stop("x must be a data.table")
    setattr(x, "row.names", .set_row_names(nrow(x)))
    setattr(x, "class", "data.frame")
    setattr(x, "sorted", NULL)
    setattr(x, ".internal.selfref", NULL)
}
StartTime <- System$currentTimeMillis()
cat("Job started on ", format(Sys.time(), "%a %b %d %Y %X"), ".  \n", sep = "")

Job started on Thu Oct 13 2016 08:19:58 PM.  

pb <- winProgressBar(title = "Mutual Information computation", label = paste("[", format(Sys.time(), "%a %b %d %Y %X"), "] Preparing computation...", sep = ""), min = 0, max = 2216, initial = 0, width = 520)
cores <- 6
choose <- cores * 5 # 5 times per core per batch
mcl <- makeCluster(cores)
clusterExport(mcl, c("folds"))
invisible(clusterEvalQ(mcl, library("woe")))
invisible(clusterEvalQ(mcl, library("data.table")))
invisible(clusterEvalQ(mcl, library("rpart")))
registerDoParallel(cl = mcl)
# Do loop
for (i in 1:ceiling(2216 / choose)) {
  
  # Prepare parallel loop
  mini_temp <- data[, ((i - 1) * choose + 1):min((i * choose), 2216), with = FALSE]
  mini_temp$target <- target
  setDF(mini_temp)
  clusterExport(mcl, c("mini_temp"))
  
  # Parallel loop setup
  whatever <- foreach(j = 1:((min((i * choose), 2216) - ((i - 1) * choose + 1)) + 1), .combine = "rbind", .init = default, .inorder = TRUE, .noexport = c("mini_temp", "folds")) %dopar% {
    whatever <- iv.mult(df = mini_temp,
                y = "target",
                vars = colnames(mini_temp)[j],
                sql = FALSE,
                topbin = TRUE,
                tbpct = 0.0001,
                verbose = FALSE,
                rcontrol = rpart.control(minsplit = 100, cp = 0.00001, xval = folds))
    if (length(whatever) == 0) {
      whatever <- data.frame(variable = colnames(mini_temp)[j],
                             class = "",
                             pct_bin = 0,
                             good = 0,
                             bad = 0,
                             bad_rate = 0,
                             inflation = 0,
                             pct_good = 0,
                             pct_bad = 0,
                             odds = 0,
                             woe = 0,
                             miv = 0,
                             stringsAsFactors = FALSE)
    } else {
      whatever <- as.data.frame(whatever[[1]])
    }
    return(whatever)
  }
  invisible(clusterEvalQ(mcl, gc(verbose = FALSE)))
  
  # Harvest statistics
  
  # Do IV
  whatever2 <- merge(aggregate(whatever$miv, by = list(Feature = whatever$variable), FUN = sum), ig_frame[, c("Feature", "Number")], by = "Feature")
  ig_frame$IV[whatever2$Number] <- whatever2$x
  
  # Do Splits
  whatever2 <- merge(aggregate(whatever$miv, by = list(Feature = whatever$variable), FUN = length), ig_frame[, c("Feature", "Number")], by = "Feature")
  ig_frame$Splits[whatever2$Number] <- whatever2$x
  
  large_data <- rbind(large_data, whatever)
  
  cat("Batch ", i, ": ", mean(ig_frame$IV[((i - 1) * choose + 1):min((i * choose), 2216)]), " with ", mean(ig_frame$Splits[((i - 1) * choose + 1):min((i * choose), 2216)]), " splits - Found ", nrow(large_data), " rules (+", nrow(whatever), " rules). Max IV found was: ", max(ig_frame$IV[((i - 1) * choose + 1):min((i * choose), 2216)]), ".  \n", sep = "")
  gc(verbose = FALSE)
  CurrentTime <- System$currentTimeMillis()
  ETA <- ifelse(i == ceiling(2216 / choose), 0, (((2216 / choose) - i) * (CurrentTime - StartTime) / i / 1000))
  pb_title <- paste("Mutual Information computation [CPU=", sprintf("%07.2f", (CurrentTime - StartTime) / 1000), "s | ETA=", sprintf("%07.2f", ETA), "s]", sep = "")
  pb_iter <- ((CurrentTime - StartTime) / 1000) / (i * choose)
  setWinProgressBar(pb, value = i * choose, title = pb_title, label = paste("[", format(Sys.time(), "%X"), " | ", sprintf("%04.2f", pb_iter), " s/iter] Doing feature ", colnames(data)[i * choose], " (", sprintf("%04d", i * choose), " / 2216 = ", sprintf("%05.2f", 100 * (i * choose) / 2216), "%)...", sep = ""))
  
  
}

Batch 1: 0.0154465 with 12.43333 splits - Found 373 rules (+373 rules). Max IV found was: 0.0552983.  
Batch 2: 0.0247018 with 28.23333 splits - Found 1220 rules (+847 rules). Max IV found was: 0.05868686.  
Batch 3: 0.01565962 with 28.8 splits - Found 2084 rules (+864 rules). Max IV found was: 0.04721482.  
Batch 4: 0.02412576 with 69.13333 splits - Found 4158 rules (+2074 rules). Max IV found was: 0.06945033.  
Batch 5: 0.01582699 with 67 splits - Found 6168 rules (+2010 rules). Max IV found was: 0.05667931.  
Batch 6: 0.01183151 with 57.03333 splits - Found 7879 rules (+1711 rules). Max IV found was: 0.03446974.  
Batch 7: 0.002596302 with 27.2 splits - Found 8695 rules (+816 rules). Max IV found was: 0.007706566.  
Batch 8: 0.001305062 with 24.46667 splits - Found 9429 rules (+734 rules). Max IV found was: 0.005127166.  
Batch 9: 0.003844629 with 48.06667 splits - Found 10871 rules (+1442 rules). Max IV found was: 0.01280566.  
Batch 10: 0.002558682 with 24.7 splits - Found 11612 rules (+741 rules). Max IV found was: 0.007548303.  
Batch 11: 0.01325514 with 49.4 splits - Found 13094 rules (+1482 rules). Max IV found was: 0.0787944.  
Batch 12: 0.04184444 with 72.06667 splits - Found 15256 rules (+2162 rules). Max IV found was: 0.08536611.  
Batch 13: 0.03745128 with 87.76667 splits - Found 17889 rules (+2633 rules). Max IV found was: 0.08519817.  
Batch 14: 0.02331317 with 59.53333 splits - Found 19675 rules (+1786 rules). Max IV found was: 0.1888925.  
Batch 15: 0.004385769 with 47.7 splits - Found 21106 rules (+1431 rules). Max IV found was: 0.01092735.  
Batch 16: 0.006184872 with 56.96667 splits - Found 22815 rules (+1709 rules). Max IV found was: 0.02319138.  
Batch 17: 0.000613954 with 9.1 splits - Found 23088 rules (+273 rules). Max IV found was: 0.003282883.  
Batch 18: 0.0002594504 with 10.93333 splits - Found 23416 rules (+328 rules). Max IV found was: 0.003060614.  
Batch 19: 0.0003178432 with 4.9 splits - Found 23563 rules (+147 rules). Max IV found was: 0.002242485.  
Batch 20: 2.155617e-05 with 3.4 splits - Found 23665 rules (+102 rules). Max IV found was: 0.0005943852.  
Batch 21: 0.0006674818 with 14.13333 splits - Found 24089 rules (+424 rules). Max IV found was: 0.003316824.  
Batch 22: 0.0005191635 with 16.96667 splits - Found 24598 rules (+509 rules). Max IV found was: 0.002015877.  
Batch 23: 0.02090894 with 46.16667 splits - Found 25983 rules (+1385 rules). Max IV found was: 0.08340856.  
Batch 24: 0.01697376 with 50.4 splits - Found 27495 rules (+1512 rules). Max IV found was: 0.06174175.  
Batch 25: 0.0845686 with 15.6 splits - Found 27963 rules (+468 rules). Max IV found was: 0.1879031.  
Batch 26: 0.01331199 with 3.6 splits - Found 28071 rules (+108 rules). Max IV found was: 0.06149098.  
Batch 27: 0.01865425 with 4.466667 splits - Found 28205 rules (+134 rules). Max IV found was: 0.1640065.  
Batch 28: 0.07968504 with 4.133333 splits - Found 28329 rules (+124 rules). Max IV found was: 0.2856957.  
Batch 29: 0.006640682 with 38.06667 splits - Found 29471 rules (+1142 rules). Max IV found was: 0.02194735.  
Batch 30: 0.005559084 with 38.63333 splits - Found 30630 rules (+1159 rules). Max IV found was: 0.02037056.  
Batch 31: 0.001645852 with 9.366667 splits - Found 30911 rules (+281 rules). Max IV found was: 0.01401855.  
Batch 32: 2.752827e-05 with 1.133333 splits - Found 30945 rules (+34 rules). Max IV found was: 0.0001463404.  
Batch 33: 0.0001115851 with 1.033333 splits - Found 30976 rules (+31 rules). Max IV found was: 0.002565435.  
Batch 34: 0.0004295566 with 1.266667 splits - Found 31014 rules (+38 rules). Max IV found was: 0.002138272.  
Batch 35: 0.007773453 with 1.433333 splits - Found 31057 rules (+43 rules). Max IV found was: 0.02702745.  
Batch 36: 7.379473e-05 with 1.033333 splits - Found 31088 rules (+31 rules). Max IV found was: 0.0006908366.  
Batch 37: 0.001977097 with 1.2 splits - Found 31124 rules (+36 rules). Max IV found was: 0.01425203.  
Batch 38: 0.02751318 with 1.433333 splits - Found 31167 rules (+43 rules). Max IV found was: 0.4926713.  
Batch 39: 0.1379648 with 273.2333 splits - Found 39364 rules (+8197 rules). Max IV found was: 0.57725.  
Batch 40: 0.05938486 with 223.6333 splits - Found 46073 rules (+6709 rules). Max IV found was: 0.3370565.  
Batch 41: 0.03602823 with 129.7333 splits - Found 49965 rules (+3892 rules). Max IV found was: 0.136041.  
Batch 42: 0.007719532 with 53.93333 splits - Found 51583 rules (+1618 rules). Max IV found was: 0.1369454.  
Batch 43: 0.2448408 with 195.0667 splits - Found 57435 rules (+5852 rules). Max IV found was: 1.032843.  
Batch 44: 0.1329912 with 194.4 splits - Found 63267 rules (+5832 rules). Max IV found was: 1.066158.  
Batch 45: 0.0001513323 with 1 splits - Found 63297 rules (+30 rules). Max IV found was: 0.00453997.  
Batch 46: 0.008592273 with 1 splits - Found 63327 rules (+30 rules). Max IV found was: 0.02687168.  
Batch 47: 0.0004848828 with 1.033333 splits - Found 63358 rules (+31 rules). Max IV found was: 0.008784763.  
Batch 48: 0.02249145 with 1.133333 splits - Found 63392 rules (+34 rules). Max IV found was: 0.316063.  
Batch 49: 0.1573172 with 335.5667 splits - Found 73459 rules (+10067 rules). Max IV found was: 0.57725.  
Batch 50: 0.02712369 with 101.7 splits - Found 76510 rules (+3051 rules). Max IV found was: 0.1336369.  
Batch 51: 0.007874745 with 63.53333 splits - Found 78416 rules (+1906 rules). Max IV found was: 0.1369454.  
Batch 52: 0.1866265 with 179.6333 splits - Found 83805 rules (+5389 rules). Max IV found was: 1.066158.  
Batch 53: 0.1477571 with 322.0667 splits - Found 93467 rules (+9662 rules). Max IV found was: 0.57725.  
Batch 54: 0.02167264 with 94.36667 splits - Found 96298 rules (+2831 rules). Max IV found was: 0.3370565.  
Batch 55: 0.02860292 with 119.1333 splits - Found 99872 rules (+3574 rules). Max IV found was: 0.1369454.  
Batch 56: 0.06534473 with 74.36667 splits - Found 102103 rules (+2231 rules). Max IV found was: 0.8584556.  
Batch 57: 0.2584039 with 328.7667 splits - Found 111966 rules (+9863 rules). Max IV found was: 1.066158.  
Batch 58: 0.05401922 with 184.0333 splits - Found 117487 rules (+5521 rules). Max IV found was: 0.3370565.  
Batch 59: 0.03316001 with 136.6333 splits - Found 121586 rules (+4099 rules). Max IV found was: 0.1369454.  
Batch 60: 0.001645289 with 19.63333 splits - Found 122175 rules (+589 rules). Max IV found was: 0.006712436.  
Batch 61: 0.2236541 with 222.7333 splits - Found 128857 rules (+6682 rules). Max IV found was: 1.066158.  
Batch 62: 0.005863716 with 11.63333 splits - Found 129206 rules (+349 rules). Max IV found was: 0.02036483.  
Batch 63: 0.01199258 with 12.26667 splits - Found 129574 rules (+368 rules). Max IV found was: 0.03640654.  
Batch 64: 0.002771426 with 7.366667 splits - Found 129795 rules (+221 rules). Max IV found was: 0.01063805.  
Batch 65: 0.05406482 with 95.53333 splits - Found 132661 rules (+2866 rules). Max IV found was: 0.3632893.  
Batch 66: 0.01328544 with 50.66667 splits - Found 134181 rules (+1520 rules). Max IV found was: 0.08681855.  
Batch 67: 0.004392374 with 25.1 splits - Found 134934 rules (+753 rules). Max IV found was: 0.03496181.  
Batch 68: 0.007772502 with 17.76667 splits - Found 135467 rules (+533 rules). Max IV found was: 0.03088115.  
Batch 69: 0.0384596 with 26 splits - Found 136247 rules (+780 rules). Max IV found was: 0.4160906.  
Batch 70: 0.006765668 with 14.9 splits - Found 136694 rules (+447 rules). Max IV found was: 0.1606605.  
Batch 71: 0.0003422376 with 1 splits - Found 136724 rules (+30 rules). Max IV found was: 0.001665384.  
Batch 72: 0.008600526 with 1 splits - Found 136754 rules (+30 rules). Max IV found was: 0.02687168.  
Batch 73: 0.001286627 with 1 splits - Found 136784 rules (+30 rules). Max IV found was: 0.01386439.  
Batch 74: 0.03094415 with 6.346154 splits - Found 136949 rules (+165 rules). Max IV found was: 0.316063.  

registerDoSEQ()
stopCluster(mcl)
closeAllConnections()
invisible(close(pb))
gc(verbose = TRUE)

Garbage collection 165 = 37+8+120 (level 2) ... 
103.1 Mbytes of cons cells used (60%)
18699.2 Mbytes of vectors used (63%)

             used    (Mb) gc trigger    (Mb)   max used    (Mb)
Ncells    1929088   103.1    3205452   171.2    2637877   140.9
Vcells 2450938346 18699.2 3910687954 29836.2 2572595015 19627.4

cat("Computation time: ", sprintf("%07.2f", (System$currentTimeMillis() - StartTime) / 1000), "s.  \n", sep = "")

Computation time: 6280.98s.  

cat("Job over on ", format(Sys.time(), "%a %b %d %Y %X"), ".  \n", sep = "")

Job over on Thu Oct 13 2016 10:04:39 PM.  

Store splits

mass_data <- large_data
mass_data$class <- as.character(mass_data$class)
fwrite(mass_data, "E:/Laurae/NumericCMI_exact_best_grid/IG_IV_splits.csv")

Setup feature names properly for Numeric

all_frame <- ig_frame
all_frame$RankIV <- (nrow(all_frame) + 1) - rank(all_frame$IV, ties.method = "max")
all_frame$Number <- NULL
gc(verbose = TRUE)

Garbage collection 166 = 37+8+121 (level 2) ... 
103.1 Mbytes of cons cells used (60%)
18699.3 Mbytes of vectors used (63%)

             used    (Mb) gc trigger    (Mb)   max used    (Mb)
Ncells    1929641   103.1    3205452   171.2    2711633   144.9
Vcells 2450943581 18699.3 3910687954 29836.2 2572595015 19627.4

fwrite(all_frame, "E:/Laurae/NumericCMI_exact_best_grid/IG_IV_scores.csv")

Pretty print DT Information Gain

Leak reference: IV = 2.843399 (super large!!!) - 3 splits: [-Inf, -1.5), [-1.5, 164), [164, +Inf]

datatable(all_frame,
          filter = "top",
          class = "cell-border stripe",
          plugins = "natural",
          extensions = c("AutoFill",
                         #"Buttons",
                         "ColReorder",
                         "KeyTable",
                         "Responsive",
                         "RowReorder"),
          options = list(style = "bootstrap",
                         pageLength = 20,
                         lengthMenu = c(5, 10, 15, 20, 25, 50, 100, 250, 500, 1000, 2500),
                         order = list(list(7, "desc")),
                         autofill = TRUE,
                         #dom = "Bfrtip",
                         #buttons = c("copy", "csv", "excel", "pdf", "print"),
                         colReorder = TRUE,
                         keys = TRUE,
                         rowReorder = TRUE,
                         searchHighlight = TRUE,
                         search = list(regex = TRUE, caseInsensitive = FALSE))
          ) %>% formatStyle('IG',
                             background = styleColorBar(range(all_frame$IG, na.rm = TRUE, finite = TRUE), 'lightgreen'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('IV',
                             background = styleColorBar(range(all_frame$IV, na.rm = TRUE, finite = TRUE), 'pink'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('Splits',
                             background = styleColorBar(range(all_frame$Splits, na.rm = TRUE, finite = TRUE), 'yellow'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('NotMissing',
                             background = styleColorBar(c(0, 1), 'lightgrey'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('Levels',
                             background = styleColorBar(range(all_frame$Levels, na.rm = TRUE, finite = TRUE), 'lightgrey'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatRound(columns = c("IG"),
                            digits = 8) %>%
                formatRound(columns = c("IV"),
                            digits = 8) %>%
                formatPercentage(columns = c("NotMissing"),
                                 digits = 4)

Look at the splits

Table:

• variable = Variable
• class = Split
• pct_bin = Percentage of samples included in the split
• good = Count of non-defect in the split
• bad = Count of defects in the split
• bad_rate = Relative count of defects in the split vs population
• inflation = Inflation rate of sample count on rates (thresholded)
• pct_good = Percentage of non-defect in the split
• pct_bad = Percentage of defect in the split
• odds = Odds ratio of the split
• woe = Weight of Evidence of the split
• miv = Information Value of the split

datatable(mass_data[mass_data$inflation >= 1, ],
          filter = "top",
          class = "cell-border stripe",
          plugins = "natural",
          extensions = c("AutoFill",
                         #"Buttons",
                         "ColReorder",
                         "KeyTable",
                         "Responsive",
                         "RowReorder"),
          options = list(style = "bootstrap",
                         pageLength = 20,
                         lengthMenu = c(5, 10, 15, 20, 25, 50, 100, 250, 500, 1000, 2500),
                         order = list(list(13, "desc")),
                         autofill = TRUE,
                         #dom = "Bfrtip",
                         #buttons = c("copy", "csv", "excel", "pdf", "print"),
                         colReorder = TRUE,
                         keys = TRUE,
                         rowReorder = TRUE,
                         searchHighlight = TRUE,
                         search = list(regex = TRUE, caseInsensitive = FALSE))
          ) %>% formatStyle('woe',
                             background = styleColorBar(range(mass_data$woe, na.rm = TRUE, finite = TRUE), 'lightgreen'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('miv',
                             background = styleColorBar(range(mass_data$miv, na.rm = TRUE, finite = TRUE), 'pink'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('good',
                             background = styleColorBar(range(mass_data$good, na.rm = TRUE, finite = TRUE), 'yellow'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('bad',
                             background = styleColorBar(range(mass_data$bad, na.rm = TRUE, finite = TRUE), 'orange'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('pct_good',
                             background = styleColorBar(c(0, 1), 'yellow'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('pct_bad',
                             background = styleColorBar(c(0, 1), 'orange'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('bad_rate',
                             background = styleColorBar(c(0, 1), 'orange'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('pct_bin',
                             background = styleColorBar(c(0, 1), 'lightgrey'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatStyle('inflation',
                             background = styleColorBar(range(mass_data$inflation, na.rm = TRUE, finite = TRUE), 'lightgrey'),
                             backgroundSize = '100% 90%',
                             backgroundRepeat = 'no-repeat',
                             backgroundPosition = 'center') %>%
                formatRound(columns = c("woe"),
                            digits = 6) %>%
                formatRound(columns = c("miv"),
                            digits = 6) %>%
                formatRound(columns = c("odds"),
                            digits = 6) %>%
                formatRound(columns = c("inflation"),
                            digits = 6) %>%
                formatPercentage(columns = c("pct_bin"),
                                 digits = 4) %>%
                formatPercentage(columns = c("bad_rate"),
                                 digits = 4) %>%
                formatPercentage(columns = c("pct_good"),
                                 digits = 4) %>%
                formatPercentage(columns = c("pct_bad"),
                                 digits = 4)

It seems your data is too big for client-side DataTables. You may consider server-side processing: http://rstudio.github.io/DT/server.htmlIt seems your data is too big for client-side DataTables. You may consider server-side processing: http://rstudio.github.io/DT/server.html